PROJECT SUMMARY In 2003, we showed that the anterior pituitary hormone thyrotropin (a.k.a. TSH), hitherto known to promote thyroid hormone secretion, is a potent direct regulator of bone mass (Abe et al, Cell, 2003, PMID: 14567913)1. This finding underscored a potential role for low circulating TSH levels in causing the bone loss that has been recognized in patients with hyperthyroidism for over a century2, and, by tradition, has been attributed solely to thyroid hormone excess. We found instead that Tsh receptor-deficient Tshr-/- mice had profound osteoporosis, even when rendered euthyroid1. Importantly, we showed more recently that bone loss in Tshr-/- mice rendered hyperthyroid significantly exceeded that in wild type hyperthyroid mice (Baliram et al, J Clin Invest, 2012, PMID: 22996689)3 ? this finding not only confirmed a direct permissive action of Tshr deficiency on bone, but also buttressed multiple clinical studies showing a tight and highly reproducible correlation between low TSH levels, bone loss, and a high fracture risk in cohorts of hyperthyroid patients worldwide4-24. Furthermore, we found that the osteoclastogenic cytokine, Tnf?, was grossly elevated in Tshr-/- mice, and that its genetic deletion rescued the skeletal phenotype of Tshr deficiency (Hase et al, PNAS, 2006, PMID: 16908863; Sun et al, PNAS, 2013, PMID: 23716650)25,26. This led to the question: which cell ? osteoblast or osteoclast ? drives the effect, and which of the two Tnf receptors, Tnfrsf1a or Tnfrsf1b, mediate the action of Tnf? in Tshr deficiency? Specific Aim 1 will study mice in which the Tshr is deleted selectively in osteoblasts or osteoclasts, as well as double mutants in which both the Tshr and either Tnfrsf1a or Tnfrsf1b are deleted. Complementary co-culture experiments will determine if osteoblastic Tnf? mediates the hyper-resorption in Tshr-/- mice. A second corpus of data, confirmed by other groups27-33, showed that Tsh displays both anti- resorptive and anabolic actions1,34-37. For example, intermittent low dose Tsh injections restored the lost bone 7 months post-ovariectomy, importantly without elevating T4 levels (Sun et al, PNAS, 2008, PMID: 18332426)37. A follow-up question thus arises: is the Tshr a druggable target? Towards finding an answer, we will utilize both genetic and pharmacological approaches. In Specific Aim 2, we will examine whether high Tsh levels are anabolic using mice in which the expression of dominant-negative Tr??337 in the thyrotrope clamps Tsh at ~30-fold higher circulating levels. In Specific Aim 3, we will study the effects of a small molecule activator of the Tshr, MS438, which, we have found, binds Tshrs selectively and with a nanomolar affinity (Latif et al, Thyroid, 2015, PMID: 25333622)38. We also find that MS438 displays pro-osteoblastic and anti-osteoclastic actions in vitro, and does not elevate serum T4. We will thus inject mice with MS438 immediately (?prevention?) or 7-months following (?restoration?) ovariectomy to determine if it can prevent bone loss and/or restore the lost bone. Together, these studies should not only allow an in-depth understanding of Tsh action on bone, but also provide proof-of-concept for a new approach that targets the skeletal Tshr.